Flow cytometry is a widely used tool for high-throughput quantitative analysis of cell populations and intracellular content. Signals in flow cytometry arise from electrical impedance, forward or side light scattering, and fluorescence. Scattering and electrical impedance provide granularity and size/volume information, but with no chemical specificity. Fluorescent labeling acts as the primary approach for cellular analysis in flow cytometry. Nevertheless, fluorescent tags are not applicable to all cases, especially small molecules (e.g. drugs) for which labeling may significantly perturb their properties. The current application aims to fill this gap through the development of a multichannel spectral flow cytometer using stimulate Raman scattering (SRS) signal from inherent molecular vibration. The stimulated Raman scattering overcomes the low signal level in spontaneous Raman scattering. It measures the light-matter energy transfer and is therefore free of the nonresonant background encountered in coherent anti-Stokes Raman scattering. Moreover, as a nonlinear optical process it is inherently phase matched, permitting a weakly focused collinear beam geometry that is compatible with high-speed detection of flowing objects. The planned instrumentation contains two specific aims. The first is to build a single-frequency SRS flow cytometer using a femtosecond laser source. The second is to build a multichannel SRS flow cytometer by multiplex detection of spectrally dispersed SRS signals. Based on the large signal level, we expect to reach the speed of 10,000 cells per second. Performance of the label-free spectral cytometer will be tested through quantitation of fat storage in adipocytes and of drug uptake by cancer cells.